Research Scientist — Agentic Systems & AI Safety

White Circle, a Paris-based AI safety company, is seeking a research scientist to build autonomous, large-scale environments that push LLM agents to failure and study how they actually break.

You will run end-to-end experiments against external APIs and our models, develop realistic multi-agent environments, and surface failure modes to inform both internal models and public writeups. A strong track record in empirical AI research is required.